ML18065A122

Provides Addl Info Re TS Change Request Re Diesel Generator Testing

ML18065A122
Person / Time
Site:	Palisades
Issue date:	09/21/1995
From:	Smedley R CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML18065A123	List: ML18065A122 ML18065A124
References
NUDOCS 9509260116
Download: ML18065A122 (7)

Text

~* '

POWERING MICHIGAN'S PROGRESS Palisades Nuclear Plant: 27780 Blue Star Memorial Highway, Covert, Ml 49043 September 21, 1995 U S Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 DOCKET 50-255 - LICENSE DPR PALISADES PLANT TECHNICAL SPECIFICATIONS CHANGE REQUEST - DIESEL GENERATOR TESTING ADDITIONAL INFORMATION On May 1, 1995, Consumers Power Company requested changes to the Palisades Plant Technical Specifications for Diesel Generator {DG) testing. Those changes are intended to: 1) clarify the monthly starting test requirement, 2) add additional requirements to the monthly load test, and 3) clarify a requirement relating to the maximum DG electric loading.

1) The existing starting test requirement is:

Each diesel generator shall be manually started each month and demonstrated to be ready for loading within 10 seconds. The signal initiated to start the diesel shall be varied from one test to another to verify that A and B starting circuits are operable.

The proposed starting test requirement is:

Verify, each 31 days, that each DG starts from standby conditions and achieves ~ 2000 V in ~ 10 seconds, steady state voltage

~ 2280 and~ 2520 V, and frequency~ 59.5 and~ 61.2 Hz.

2) The existing loading test requirement is:

The generator shall be synchronized from the control room, and loaded to 2400 +/-100 kW.

9509260116 950921 PDR ADDCK 05000255 P _ PDR A~~COMPANY

2 The proposed loading test requirement is:

Verify, each 31 days, that each DG operates for ~ 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />; for

~ 15 minutes loaded above the peak accident loading, and for the remainder of the test loaded ~ 2300 and ~ 2500 kW.

3) The existing maximum loading requirement is:

Diesel generator electric loads shall not be increased beyond the continuous rating of 750 amp at 2400 volts.

The proposed maximum loading requirement is:

Verify, by analytical means, each 18 months that diesel generator steady state automatically connected electric loads do not exceed the continuous rating of 750 amp at 2400 volts.

During the review of the subject Technical Specifications change request several questions were posed by the NRR reviewer, Mr. Amar Pal. The answers to those questions are provided below:

la) What is the basis for the 2000 volts specified as the timing criterion of proposed surveillance requirement 4.7.1.a.l?

The specified 2000 Volts is the setpoint of voltage sensing relays which initiate Diesel Generator (DG) breaker closure during an automatic starting sequence. The 2000 volt setting was chosen, as part of the original Bechtel design, to indicate that the DG had attained a successful start. With a fixed excitation, the voltage is roughly proportional to speed; therefore, attainment of 2000 volts indicates that the engine has started, the engine is accelerating, and the voltage regulator is functioning. While no analytical basis for this setting selection is available, testing conducted during pre-operational testing in 1970 and 1971, and each refueling thereafter under SR 4.7.1.b, has demonstrated that this setting is adequate.

Those refueling interval tests are initiated by simulating an undervoltage on the 2400 volt bus simultaneously with a safety injection signal whi£h initiates separation of the safeguards bus from off-site power, 2400 and 480 Volt load shedding, starting of the DG, closing of the DG breaker, and starting and accelerating of sequenced accident loads.

Even though the safeguards pumps are not providing full accident flows during the tests, the re-energizing of the permanently connected loads and the acceleratiflg of the sequenced accident loads provide the same initial loading during the test as they would during an accident. No difficulties with the setting of the DG voltage sensing relays have been evident in these tests.

These DG voltage sensing relays actuate to provide contact closure, when their setting is reached, on any DG start; they actuate the breaker closure only during an automatic start sequence initiated by a

3 loss of bus voltage. The actuation of these relays is used, during monthly testing, to determine when the DG is "ready for loading". The proposed wording was chosen to clarify the testing requirement, rather than to change it.

The proposed test requirement stipulates a steady state voltage requirement which has been demonstrated analytically to be acceptable.

lb) With reference to surveillance requirement 4.7.1 b, which motors are actually selected for this test?

To determine the testing philosophy for this long standing surveillance requirement, the test procedure for the left safeguards train testing (RT-BC, Rev 4) was reviewed and compared to the listing of sequenced loads (Logic Drawtng E-17, Sheet 4, Rev 10) to determine which automatically started motors were tested, which were tested by actual operation, and which were tested by actuation of the circuit breaker in the "TEST" position. All sequenced loads are included in the test; the only sequenced loads which are tested with breakers in the test position are the HPSI pump, charging pump, and containment spray pump.

While this test does not demonstrate full accident loading, it does demonstrate the adequacy of the questioned 2000 volt setting of the relays which initiate DG breaker closing. Satisfactory operation of the permanently connected loads and the motor operated valves, which are the first loads actuated by the sequencer, do provide actual accident loads for the first few seconds of the test.

le) How quickly, after attaining 2000 Volts, do we reach steady state voltage?

Oscillograph recordings made during special tests show that we reach 2400 volts within about 1/2 second after reaching 2000 volts. During monthly testing, when no loads are being sequenced onto the DG, oscillations dampen out within I or 2 additional seconds.

On an automatic start sequence, the voltage sensing relays actuate closing the DG breaker. Breaker closing energizes the permanently connected loads, 172 kW on DG 1-1 and 222 kW on DG 1-2, and starts the automatic loading sequence. The first sequenced loads are the motor operated safety injection valves. These motor operated valves add about 30 kW. The first large motor is started at 2 seconds.

Id) What is the minimum DG terminal voltage necessary to assure adequate voltage for the 2400 &480 volt safeguards loads?

We do not have analysis to provide that specific information.

However, we have analysis to show that there is adequate voltage on the 2400 and 480 volt buses (the 480 loads being limiting) when

4 voltage reaches 91% of 2400 volts (21S4 volts). The proposed acceptance criterion is 95% of the nominal DG rating of 2400 volts (22SO). The cable runs between DG and bus are short, on the order of 100 feet, and cause no significant voltage loss.

2a) Discuss specifying power factor for load testing at least once per cycle.

This change request does not propose adding a requirement for testing with a specified power factor range. Palisades has committed to submit a request to convert our Technical Specifications to Improved Standard Technical Specifications. Standard Technical Specifications for CE Plants, NUREG 1432, contains an lS month surveillance requirement to test each DG for ~ 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> at specified loading and power factor. Testing with a specified power factor range will be evaluated during our conversion to STS.

3a) With respect to TS 4.7.1.d; The NRC reviewer's estimates indicate that, upon starting Auxiliary Feedwater Pump PS-A, the sequenced load would exceed the 750 Amp limit. Provide details of accident loading, both steady state and transient.

The attachment provides the limiting accident loading for each DG in tabular and graphical form.

The peak running loads which are automatically sequenced (those loads shown in time block zero of the attached table) are 2556 kW (707 Amps) for DG 1-1 and 2471 kW (693 Amps) for DG 1-2. The peak running loads, including loads which are automatically sequenced and loads which might be added by the operators, are 25S4 kW for DG 1-1 and 2559 for DG 1-2.

3b) Does transient loading exceed the 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> rating of either DG? Calculations by the NRC reviewer using a starting power factor of 0.2, an efficiency of 0.91, and a starting current of 6.5 times running current, show that the transient loading exceeds the two hour rating of the DG.

No. The peak starting load occurs on DG 1-1 which would be carrying 2202 kW at the start of the final sequenced large load, 450 HP AFW pump PS-A. That starting load is 397 kW, for a maximum transient load of 2599 kW. The 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> rating is 2750 kW.

The manufacturer's curves for PS-A provide a locked rotor current of 630 Amps at its rating of 2300 Volts and a locked rotor power factor of 0.1S7. The starting load, at rated voltage, would be 469 Kw.

During the start of PS-A, however, DG voltage is computed to be 2041 Volts, resulting in a starting current of 559 Amps and a starting load of 397 kW (modeled as a constant impedance with a pf of 0.2).

5 3c) Do we consider any DG "de-rating factors" based on air intake temperature?

No. The DGs are rated at sea level and ll0°F; when standard sea level pressure, 29.92" Hg, is corrected for the installed elevation, 590' MSL, the temperature for equivalent atmospheric density is 98.7°F.

Palisades FSAR section 2.5.1 states "On the basis of a study of US Weather Bureau (USWB) records for stations reasonably close to the area, the extreme 10-year minimum temperature would be about -25°F.

Because of the proximity of the cool lake, the extreme 10-year maximum temperature would be a moderate 95°F. 11 The ASHRAE handbook lists weather data for the Benton Harbor airport, about 15 miles south and about I mile from Lake Michigan, which indicates that the local dry bulb temperature will exceed 91°F only 1% of the time. Therefore, derating of the engines due to intake air temperature is not necessary.

3d) With respect to our calculated accident electrical load profile:

a. Does the accident load profile calculation consider transformer and cable losses?

Yes. The accident load profile calculation includes calculated losses for transformers and cables.

b. What are the bases for assumed power factors and efficiencies?

The power factors and efficiencies assumed for the accident load profile calculations are taken from manufacturer's curves for each large sequenced motor (some smaller loads are modeled as lumped 1oads).

c. What load is assumed for the battery chargers?

The load assumed for the battery chargers is 90 amps. They are 480 Volt loads. That is the AC load at their full capacity.

d. Does the accident load profile calculation use actual or rated loads?

The accident load profile calculation uses calculated loads determined from analytical modeling of the fluid systems. It does not simply use name plate ratings as accident loads.

3e) Have we demonstrated that each DG can actually carry its design load (2750 kW)? What duration has been demonstrated for the continuous rating?

Yes. Within the past year, each DG has been tested at or above the 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> rating of 2750 kW. It is current testing procedure to test each

6 DG at or above peak accident loading for ~ 15 minutes during each monthly surveillance test. It is not, however, general practice to routinely load the DG above the 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> rating of 2750 kW.

SUMMARY

OF COMMITMENTS This letter contains no new commitments or revisions to former commitments.

\-l-W o::;:_ SJ ~

Richard WSmedley Manager, Licensing CC Administrator, Region III, USNRC Resident Inspector, Palisades Attachment

ATTACHMENT 1 CONSUMERS POWER COMPANY PALISADES PLANT DOCKET 50-255 TECHNICAL SPECIFICATION CHANGE REQUEST - DIESEL GENERATOR TESTING ADDITIONAL INFORMATION Diesel Generator Accident Loading Information 3 Pages